Automatic generation of narratives from data using communication goals and narrative analytics

Patent metadata

Field	Value
Publication number	US-9720899-B1
Application number	US-201414521264-A
Country	US
Kind code	B1
Filing date	Oct 22, 2014
Priority date	Jan 7, 2011
Publication date	Aug 1, 2017
Grant date	Aug 1, 2017

Abstract

The exemplary embodiments described herein are related to techniques for automatically generating narratives about data based on communication goal data structures that are associated with configurable content blocks. The use of such communication goal data structures facilitates modes of operation whereby narratives can be generated in real-time and/or interactive manners.

First claim

What is claimed is:

1. A method comprising:
a processor selecting a communication goal data structure, the selected communication goal data structure comprising (1) first data that is indicative of a content block data structure associated with the selected communication goal data structure, and (2) second data that is indicative of a plurality of communication goal parameters whose values are variable, the associated content block data structure comprising a parameterized narrative analytics model, the parameterized narrative analytics model configured to specify how a data set is to be analyzed in order that a narrative be automatically generated based upon the data set, and wherein the parameterized narrative analytics model includes a plurality of angles for testing against the data set to identify an angle that is deemed to accurately characterize the data set, the angles having a plurality of applicability conditions, wherein the selected communication goal data structure comprises a data structure configured to specify how a status for a subject is to be explained, the communication goal parameters that are associated with the explain subject status communication goal data structure comprising:
a subject metric parameter;
a time frame parameter; and
a driver parameter for the subject metric parameter;

the processor configuring the content block data structure associated with the selected communication goal data structure based on the communication goal parameters associated with the selected communication goal data structure, wherein the configuring step comprises:
the processor receiving input;
the processor defining a plurality of values for the communication goal parameters based on the received input; and
the processor defining the content block data structure associated with the selected communication goal data structure as a function of the defined values for the communication goal parameters;

the processor storing the configured content block data structure in memory to thereby enable an automated generation of the narrative based on the configured content block data structure as applied to the data set;
the processor receiving the data set;
the processor processing the received data set based on the configured content block data structure; and
the processor automatically generating a narrative about the data set based on the processing, wherein the generated narrative is tailored to the explain subject status communication goal.

2. The method of claim 1 wherein the driver parameter comprises a plurality of the driver parameters.

3. The method of claim 2 wherein the plurality of driver parameters comprises a contributing driver and an inhibiting driver.

4. The method of claim 1 wherein the input receiving step comprises the processor receiving the input from a user through a user interface, the user interface comprising a plurality of fields for user input of values for the communication goal parameters associated with the selected communication goal data structure.

5. The method of claim 4 further comprising:
the user interface providing the user of a preview of an output narrative from the data set based on the defined communication goal parameter values.

6. The method of claim 1 wherein the communication goal parameters that are associated with the selected communication goal data structure further comprise:
an information density parameter.

7. The method of claim 1 wherein the communication goal parameters that are associated with the selected communication goal data structure further comprise:
a tone parameter.

8. The method of claim 1 wherein the selected communication goal data structure further comprises third data that is indicative of another communication goal data structure that is associated with the selected communication goal data structure.

9. The method of claim 1 wherein the parameterized narrative analytics model further specifies a plurality of blueprint sets associated with the content block data structure and the angles.

10. A method comprising:
a processor selecting a communication goal data structure, the selected communication goal data structure comprising (1) first data that is indicative of a content block data structure associated with the selected communication goal data structure, (2) second data that is indicative of a plurality of communication goal parameters whose values are variable, and (3) third data that is indicative of another communication goal data structure that is associated with the selected communication goal data structure, the associated content block data structure comprising a parameterized narrative analytics model, the parameterized narrative analytics model configured to specify how a data set is to be analyzed in order that a narrative be automatically generated based upon the data set, wherein the parameterized narrative analytics model includes a plurality of angles for testing against the data set to identify an angle that is deemed to accurately characterize the data set, the angles having a plurality of applicability conditions, and
the processor configuring the content block data structure associated with the selected communication goal data structure based on the communication goal parameters associated with the selected communication goal data structure;
the processor storing the configured content block data structure in memory to thereby enable an automated generation of the narrative based on the configured content block data structure as applied to the data set;
the processor receiving the data set;
the processor processing the received data set based on the configured content block data structure; and
the processor automatically generating a narrative about the data set based on the processing, wherein the generated narrative is tailored to a communication goal associated with the selected communication goal data structure;
wherein the another communication goal data structure comprises (1) first data that is indicative of another content block data structure that is associated with the another communication goal data structure, and (2) second data that is indicative of another plurality of communication goal parameters whose values are variable, the associated another content block data structure comprising another parameterized narrative analytics model, the another parameterized narrative analytics model configured to specify how another narrative is to be automatically generated from a data set about which the another narrative is to be automatically generated, the another narrative analytics model including a plurality of angles for testing against the data set for the another parameterized narrative analytics model to identify an angle that is deemed to accurately characterize the data set for the another parameterized narrative analytics model, the angles having a plurality of applicability conditions, the method further comprising:
the processor identifying the another communication goal data structure that is associated with the selected communication goal data structure;
the processor configuring the another content block data structure associated with the identified another communication goal data structure based on the another communication goal parameters associated with the identified another communication goal data structure; and
the processor storing the configured another content block data structure in memory to thereby enable an automated generation of the another narrative based on the configured another content block data structure as applied to the data set for the another parameterized narrative analytics model.

11. The method of claim 10 wherein the configurin

Assignees

• Narrative Science Inc

Inventors

• Birnbaum Lawrence
• Hammond Kristian J
• Nichols Nathan Drew
• Paley Andrew R
• Murata Shingo
• Opulencia Pia

Classifications

• G06F40/253

  Grammatical analysis; Style critique · CPC title

• G06Q10/0639

  Performance analysis of employees; Performance analysis of enterprise or organisation operations · CPC title

• G06Q30/0201

  Market modelling; Market analysis; Collecting market data · CPC title

• G06N20/00

  Machine learning · CPC title

• G06Q10/06398

  Performance of employee with respect to a job function · CPC title